Network Science for Graph Practitioners: Seeing Beyond Nodes and Edges
A Talk by Andre Franca , Orit Gal and Milan Janosov
About this Talk
Join us for an expert panel exploring the intersection of graph technology and network science. You'll learn how to move beyond simple node and edge analysis to uncover and infer deeper meanings within data. We’ll discuss the types of networks encountered across industries and what they reveal about complex systems.
The panel will highlight practical lessons, address common misconceptions, and evaluate the impact of using network science along with graph technology. We’ll also explore whether the tech industry's focus on predictions limits our understanding of systems and discuss how to incorporate network science principles into graph solutions, along with the key skills graph professionals should acquire.
Key Topics
- Network science principles
- What network science can reveal about data and networks
- Using network science alongside graph technology
Target Audience
- Graph Practitioners
- Product Managers
- Data Scientists
- Network Scientists
Goals
- Understand core network science concepts
- Explore how network science can improve graph solutions
- Avoid common mistakes and missed opportunities
Session outline:
- Introductions and Topics
- Network Science and Graphs: Two Sides of the Same Coin
- Should we consider all data as part of a system? What’s the implication of that?
- Why study networks themselves?
- How should we think about network science vs. graph theory vs. graph technology?
- What Networks Are Telling Us
- What different types of networks will people encounter? And are there common ways these networks function/operate/change?
- What kinds of insights should we use/not use network science for?
- What are the most common misconceptions about what we can learn from networks?
- Applying Network Science Thinking in Graph Technology
- How can people working on graph solutions best leverage network science tools or concepts?
- Is the current obsession with making predictions keeping the tech industry from better understanding the networks/systems we’re working on?
- Incorporating Network Science Principles
- What core network science principles should everyone incorporate into planning for graph solutions?
- Are there unique considerations for production deployments that depend on more rigorous science-based insights?
- What kind of expertise do we need for advanced analytics teams?
- Should we bring together commercial graph practitioners and network scientists into one team?
- What network science skills should graph practitioners acquire, and how?
Format
- An expert panel discussion including Milan Janosov, formerly a network scientist with Barabasi Labs and now working on geospatial; Orit Gal, a complexity lecturer and city-social researcher for urban planning; and Andre Franca, who develops causal ML tools and previously a quantum physics researcher.
- Moderated by Amy Hodler, advisor and founder of the GraphGeeks community.
- Modules of expert discussion interspersed with audience Q&A
- 2 hours
Level
- Beginning and Intermediate
Prerequisite Knowledge
- Basic understanding of graph concepts